INDEX

Explanations

certain formatting or structural elements in data, specifically those related to metadata and identifiers

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eling

-0.14

allas

-0.14

ikh

-0.13

ambda

-0.13

atis

-0.13

Vys

-0.13

rist

-0.13

WithTitle

-0.13

ings

-0.13

uke

-0.13

POSITIVE LOGITS

 Bened

0.16

ispens

0.14

.spin

0.14

Ã¼st

0.14

uve

0.14

abin

0.14

ercul

0.14

 chalk

0.14

 Revolution

0.13

 revolution

0.13

Activations Density 0.243%

certain formatting or structural elements in data, specifically those related to metadata and identifiers

No Comments

No Known Activations